Quantitative analysis of visual codewords of a protein distance matrix

The supplemental data includes all 7308 distance matrix images (grayscale, PNG format) and a Matlab script that calculates the bag of features, histograms (feature vector), and classification accuracy. 

We also provided a dataset of globular and solenoid proteins (247+105 images, grayscale, PNG format) and a Matlab script (SVM_2fold.m) to classify proteins using SVM. In summary, the supplementary material contains all the inputs needed to run Matlab scripts.

Journal:

• https://doi.org/10.1371/journal.pone.0263566